Announcement for Downloading full text filePlease respect the Copyright Act.
All digital full text dissertation and theses from this website are authorized the copyright owners. These copyrighted full-text dissertation and theses can be only used for academic, research and non-commercial purposes. Users of this website can search, read, and print for personal usage. In respect of the Copyright Act of the Republic of China, please do not reproduce, distribute, change, or edit the content of these dissertations and theses without any permission. Please do not create any work based upon a pre-existing work by reproduction, Adaptation, Distribution or other means.
URN etd-0716118-133400 Statistics This thesis had been viewed 127 times. Download 0 times. Author Chen-han Szu Author's Email Address email@example.com Department Electrical Engineering Year 2017 Semester 2 Degree Master Type of Document Master's Thesis Language zh-TW.Big5 Chinese Page Count 29 Title Achieve Early Warning For Electrical Error By Deep Learning Keyword Deep learning Neural Network Backpropagation Long Short-Term Memory Smart Electricity Grid Power Systems Electrical Error Electrical Error Power Systems Smart Electricity Grid Long Short-Term Memory Backpropagation Neural Network Deep learning Abstract Science and technology have improved our living standard, the electrical safety has been forgotten easily when the amount of electric power set a new high year by year. According to national fire analysis report made by National Fire Agency, Ministry of the Interior “Electrical fire account for 32.8% of the year 2016, and made 49 people died.” Illegal structures, overloaded power strip, decoration with inflammable materials, disrepair for long time, make potential hazards to the electrical fire.
Electrical error from live wire often leads to voltage drop and overheating of wire and device. The neutral wire error makes imbalance phases from wire, which let voltage drop from heavy load and voltage rise from light load. These situation may show a big difference from the characteristics as usual.
This research try to find a positive way of early warning utilize Backpropagation Neural Networks and Long Short-Term Memory Neural Networks for deep learning, which can protect loss of life and property from electrical error, which would complete smart electricity grid for a more stable power system.
Advisor Committee Chau-Yun Hsu - advisor
An-Yi Chen - co-chair
Wen-Chen Chu - co-chair
Files Date of Defense 2018-07-02 Date of Submission 2018-07-16